Category
Building with and for AI systems — from model integration to production ML pipelines.
Generating embeddings is the easy part. Keeping them in sync as your data changes — at scale, without downtime — is where the real engineering lives. Here's how to build the full pipeline on MongoDB Atlas.
AI engineering isn't just software engineering with a model attached. The feedback loops, failure modes, and quality signals are fundamentally different. Here's how to think about it.
Large contexts, MAX mode, Opus 4.6 Thinking — the most powerful AI coding tools are also the most expensive. After a brutal invoice, I figured out how to get 90% of the performance at 20% of the cost.
Writing code line by line is no longer the job. The job is conducting agents — knowing what to build, how to decompose it, and how to steer AI toward the right solution. The skill that matters now is engineering judgment, not typing speed.